natural-sql  by cfahlgren1

Text-to-SQL LLMs with strong performance

created 1 year ago
866 stars

Top 42.3% on sourcepulse

GitHubView on GitHub
Project Summary

NaturalSQL provides a series of state-of-the-art Text-to-SQL models, specifically designed for generating accurate SQL queries from natural language questions. It targets developers and data analysts who need to interact with databases using plain English, offering improved performance on complex queries involving subqueries and ratios.

How It Works

The project leverages large language models (LLMs) fine-tuned for the Text-to-SQL task. The NaturalSQL-7B model, in particular, demonstrates leading performance for its size, as benchmarked on the SQL-Eval framework. This approach allows for flexible and intuitive database querying without requiring explicit SQL knowledge.

Quick Start & Requirements

Highlighted Details

  • State-of-the-art performance on Text-to-SQL benchmarks.
  • NaturalSQL-7B is a leading model for its size on SQL-Eval.
  • Handles complex queries including subqueries and ratios effectively.

Maintenance & Community

The project acknowledges Defog for open-sourcing the SQL-Eval framework. Further community or maintenance details are not provided in the README.

Licensing & Compatibility

  • Code License: Apache-2.0
  • Model License: CC BY-SA 4.0
  • Compatibility: The CC BY-SA 4.0 license for the model may have implications for commercial use or integration into closed-source projects due to its share-alike clause.

Limitations & Caveats

The README does not detail specific limitations, unsupported platforms, or known bugs. The CC BY-SA 4.0 model license requires careful consideration for commercial applications.

Health Check
Last commit

1 year ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
3 stars in the last 90 days

Explore Similar Projects

Feedback? Help us improve.